As the world transitions to renewable energy, accurately forecasting wind energy generation becomes crucial for grid reliability and planning. This project, presented at the European Geosciences Union (EGU) General Assembly 2020, applies machine learning techniques to predict wind power generation using atmospheric variables from reanalysis datasets.

By engineering predictive environmental features and testing a range of ML algorithms, the study evaluates how well data-driven models can support energy forecasting at scale.

Project Highlights

  • 💨 Wind Energy Forecasting: Focused on predicting wind power generation based on environmental conditions.
  • 📊 Environmental Inputs: Used reanalysis data such as wind speed, temperature, and air pressure.
  • 🧠 ML Algorithms Tested: Evaluated models including Random Forests and Support Vector Machines.

🎤 Presented at: EGU General Assembly 2020 – Session EGU2020-8018